I am trying to replicate a tflearn model in dl4j(newbie). Some part of the architecture is replicated. There are some complications that i faced as a result i dont know if the till now done part is correct or not.
I am attaching the code ss in dl4j and the respective tflearn model link(a new user cant put more than 1 media).Please ignore the dimensions for now. If it is not against the community guideline please help.
Link to tflearn model=>https://github.com/hongzimao/pensieve/blob/1120bb173958dc9bc9f2ebff1a8fe688b6f4e93c/sim/a3c.py#L61
In the above link please refer the “define actor network”.